Contents |
lsb-test-python is a noarch package. Both 2.4.2 and 2.5 tests are bundled, and a choice is made at run-time, based on the system python version. Modules tested are only those that are common to both 2.4 and 2.5.
System Python version 2.4.2
Test was run: 20071015 10:24:54 Test Suite Name: lsb-test-python Test Suite Version: 3.1.90-3 Test Suite Architecture: noarch Total Tests Passed: 108 Total Tests Failed (including waived): 0 Total Tests Failed (excluding waived): 0
System Python version 2.4.4
Test was run: 20071015 10:34:14 Test Suite Name: lsb-test-python Test Suite Version: 3.1.90-3 Test Suite Architecture: noarch Total Tests Passed: 108 Total Tests Failed (including waived): 0 Total Tests Failed (excluding waived): 0
System Python version 2.4.3
test_codecencodings_jp|TC 21 FAIL Test was run: 20071015 10:35:59 Test Suite Name: lsb-test-python Test Suite Version: 3.1.90-3 Test Suite Architecture: noarch Total Tests Passed: 107 Total Tests Failed (including waived): 1 Total Tests Failed (excluding waived): 1
System Python version 2.5.1
Test was run: 20080107 20:12:50 Test Suite Name: lsb-test-python Test Suite Version: 3.1.95-3 Test Suite Architecture: noarch Total Tests Passed: 108 Total Tests Failed (including waived): 0 Total Tests Failed (excluding waived): 0
System Python version 2.4.0
98 tests OK.
10 tests failed:
test_StringIO test_codecs test_complex test_list test_mmap test_re
test_set test_sets test_trace test_weakref
Journal file is results/0001e/journal
test_StringIO|TC 7 FAIL test_codecs|TC 24 FAIL test_complex|TC 27 FAIL test_list|TC 60 FAIL test_mmap|TC 65 FAIL test_re|TC 78 FAIL test_sets|TC 84 FAIL test_set|TC 83 FAIL test_trace|TC 97 FAIL test_weakref|TC 105 FAIL Test was run: 20071015 11:32:10 Test Suite Name: lsb-test-python Test Suite Version: 3.1.90-3 Test Suite Architecture: noarch Total Tests Passed: 98 Total Tests Failed (including waived): 10 Total Tests Failed (excluding waived): 10
Analysis: I used 2.4.2 as the base version for the 2.4 series tests. Looking at the above tests, all but test_list have additional or changed tests in the 2.4.2 source. For test_list, the test code is identical. The failed subtest is:
test_constructor_exception_handling (test.test_list.ListTest) ... FAIL
This raises the question of whether we should be using the 2.4.0 tests, or perhaps call out 2.4.2 as the minimal Python version in the spec. A quick rebuild of the tests using 2.4.0 sources, gives 108 tests passing on CS4.0, but test_trace fails on SLES10, RHEL5, Debian4, Ubuntu 6.10, Mandriva 2007.0 (test_locale also skipped on Debian, Ubuntu).
Decision was made at the F2F to require Python 2.4.2, so these results aren't really too important.
ERROR: test_xmlcharrefreplace (test.test_codecencodings_jp.Test_EUC_JP_COMPAT)
Traceback (most recent call last):
File "/opt/lsb/test/python/Lib24/test/test_multibytecodec_support.py", line 56, in test_xmlcharrefreplace
self.assertEqual(
ValueError: unknown error handling code: xmlcharrefreplace
ERROR: test_customreplace (test.test_codecencodings_jp.Test_SJIS_COMPAT)
Traceback (most recent call last):
File "/opt/lsb/test/python/Lib24/test/test_multibytecodec_support.py", line 94, in test_customreplace
"test.xmlcharnamereplace")[0], sout)
ValueError: unknown error handling code: test.xmlcharnamereplace
ERROR: test_xmlcharrefreplace (test.test_codecencodings_jp.Test_SJIS_COMPAT)
Traceback (most recent call last):
File "/opt/lsb/test/python/Lib24/test/test_multibytecodec_support.py", line 56, in test_xmlcharrefreplace
self.assertEqual(
ValueError: unknown error handling code: xmlcharrefreplace
Ran 30 tests in 1.465s
FAILED (errors=6)
I'm not sure what these results mean (Stew)
System Python version 2.4.4c1
Test was run: 20071015 11:21:53 Test Suite Name: lsb-test-python Test Suite Version: 3.1.90-3 Test Suite Architecture: noarch Total Tests Passed: 108 Total Tests Failed (including waived): 0 Total Tests Failed (excluding waived): 0
System Python version 2.5.1
/usr/bin/python 79 FAIL Test was run: 20080117 09:42:34 Test Suite Name: lsb-test-python Test Suite Version: 3.1.95-3 Test Suite Architecture: noarch Total Tests Passed: 107 Total Tests Failed (including waived): 1 Total Tests Failed (excluding waived): 1
Test 79 (test_resource) is failing with:
test_resource test test_resource crashed -- <type 'exceptions.IOError'>: [Errno 27] File too large
A single test run doesn't give much more information:
root@ubuntu32:/opt/lsb/test/python# ./single_test -v test_resource
test_resource
True
test test_resource crashed -- <type 'exceptions.IOError'>: [Errno 27] File too large
Traceback (most recent call last):
File "./Lib25/test/regrtest.py", line 583, in runtest_inner
the_package = __import__(abstest, globals(), locals(), [])
File "/opt/lsb/test/python/Lib25/test/test_resource.py", line 42, in <module>
f.close()
IOError: [Errno 27] File too large
'test_resource' left behind file '@test'
1 test failed:
test_resource
I do see apparmor is enabled on this install. I tried both disabling apparmor and rebooting (didn't seem to stop it from loading), and just stopping the service - did not change the test results.
Also tried my 64bit Ubuntu 7.10 VM - same results
System Python version 2.4.3
Test was run: 20071015 11:23:26 Test Suite Name: lsb-test-python Test Suite Version: 3.1.90-3 Test Suite Architecture: noarch Total Tests Passed: 108 Total Tests Failed (including waived): 0 Total Tests Failed (excluding waived): 0
System Python version 2.5.1
Test was run: 20080117 10:05:00 Test Suite Name: lsb-test-python Test Suite Version: 3.1.95-3 Test Suite Architecture: noarch Total Tests Passed: 108 Total Tests Failed (including waived): 0 Total Tests Failed (excluding waived): 0